© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 55514
    Prev
    Next
    INDEX
    Explanations

    Precision/control

    np_max-act · gemini-2.0-flash

    This neuron fires on technical modifiers—adjectives and adverbs that qualify parameters or settings (e.g. “metered,” “slight,” “axial,” “temporary,” “continuous,” “direct”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    .Make
    -0.07
     oddly
    -0.07
     Bereich
    -0.07
     Tears
    -0.06
     DDS
    -0.06
     einzel
    -0.06
     Nigel
    -0.06
    stdafx
    -0.06
    ulators
    -0.06
    holders
    -0.06
    POSITIVE LOGITS
     IEnumerable
    0.07
     ekonom
    0.07
     selective
    0.06
    ."↵↵↵
    0.06
     надеж
    0.06
     ven
    0.06
    "↵↵↵
    0.06
    agne
    0.06
    两个
    0.06
     profitable
    0.06
    Activations Density 0.215%

    No Known Activations